AntagomiR-451 inhibits oxygen glucose deprivation (OGD)-induced HUVEC necrosis via activating AMPK signaling

Oxygen glucose deprivation (OGD) application in cultured human umbilical vein endothelial cells (HUVECs) mimics ischemic injuries. AntagomiR-451, the miroRNA-451 (“miR-451”) inhibitor, could activate pro-survival AMP-activated protein kinase (AMPK) signaling. In the current study, we showed that forced-expression of antagomiR-451 depleted miRNA-451 and significantly attenuated OGD-induced necrosis of HUVECs. Activation of AMPK was required for antagomiR-451-mediated pro-survival actions. AMPK inhibition, by AMPKα shRNA or dominant negative mutation, almost completely abolishedantagomiR-451-mediated HUVEC protection again OGD. Reversely, forced-activation of AMPK by exogenous expression of constructively-active AMPKα inhibited OGD-induced HUVEC necrosis. At the molecular level, antagomiR-451 expression in HUVECs inhibited OGD-induced programmed necrosis, the latter was evidenced by mitochondrial p53-cyclophilinD (Cyp-D) association, mitochondrial depolarization as well as reactive oxygen species (ROS) production and lactate dehydrogenase (LDH) breach. Together, we suggest that antagomiR-451 activates AMPK to inhibit OGD-induced programmed necrosis in HUVECs.


Introduction
Ischemic vascular diseases cause substantial injuries to the vascular valve and vascular endothelial cells, which eventually leads to damages to surrounding tissues [1,2]. The underlying signaling mechanisms are still not fully understood [1,2]. Oxygen glucose deprivation (OGD) is applied to mimic ischemic damages in vitro. Treatment of vascular endothelial cells with serve and/or sustained OGD (>1 hour) will suppress mitochondrial respiratory chain. When coupled with re-oxygenation, superoxide and other reactive oxygen species (ROS) will be produced, causing oxidative stress and cell programmed necrosis (but not apoptosis) [3,4,5].

Chemicals and antibodies
The antibodies of this study were purchased from Cell Signaling Tech (Nanjing, China). Puromycin was obtained from Sigma-Aldrich (Shanghai, China). Cell culture reagents were provided by Gibco(Shanghai, China). Transfection reagents were obtained from Invitrogen (Shanghai, China).

Cell culture
HUVECs were provided by the Hu's group at Shanghai JiaoTong University School of Medicine [19]. HUVECs were cultured in routine medium 199with 15% FCS, and endothelial cell growth supplement (Sigma) plus epidermal growth factor (EGF 10 ng/mL) [19]. All experiments using human cells (HUVECs) were approved by the Institutional Review Board (IRB, No.BS200412) and Ethics Review Board (ERB, No.2013117) of Kunming General Hospital, PLA, and were conducted according to the principles expressed in the Declaration of Helsinki.

LDH detection of cell necrosis
Cell necrosis was examined by routine lactate dehydrogenase (LDH) release assay using a commercial available two-step LDH kit (Takara, Tokyo, Japan). LDH release was calculated: LDH released in conditional medium/(LDH in conditional medium + LDH in cell lysates) × 100%.
Quantitative real-time polymerase chain reaction (qRT-PCR) assay After treatment of cells, the total RNA was extracted via the Trizol reagents (Qiagen, Shanghai, China). The concentration of total RNA was quantified via a NanoDrop Spectrophotometer (NanoDrop Technologies). The TaqMan microRNA assay system was utilized to quantify miRNA-451 expression, using the Applied Biosystems 7500 Real-Time PCR System. ΔΔ Ct method was employed to quantify mRNA expression. GAPDH was always tested as the internal control [22,23].miRNA-451 primers were provided by Dr. Lu [17].

Western blotting assay
After treatment, the cell lysis buffer (Biyuntian, Wuxi, China) was applied to achieve the protein lysates. Twenty μg lysates per sample were separated by 8-10% SDS-PAGE gels, which were then transferred onto PVDF membranes (Shanghai, China). After blocking, the implied primary and secondary antibodies were then added. Afterwards, ECL reagents (Roche, Shanghai, China) were added to detect the interested bands. Tubulin was always tested as the loading control. ImageJ software was applied to quantify the total gray of each band.

ROS assay
The intracellular ROS content was tested by the dichlorofluorescin (DCF) oxidation assay, and the detailed protocol was described previously [33]. Following the applied treatment, 10 μM DCFH-DA (Invitrogen) was added. Cells were then washed, trypsinized and resuspended in PBS. DCF fluorescence was then examined. The excitation wavelength was set at 488 nm. The DCF fluorescent OD value of treatment group was always normalized to that of untreated control group.

Lipid peroxidation assay
Thiobarbituric acid reactive substances (TBAR) level was tested to reflect lipid peroxidation, the detailed protocol was described previously [34]. Briefly, after treatment, intracellular lysates (20 μg per treatment) were mixed with acetic acid (20%) and aqueous solution of thiobarbituric acid (0.78%). After heating, the mixtures were centrifuged, and then the red pigment in the supernatant was estimated by a microplate reader. The lipid peroxide level was expressed as nM of malondialdehyde per mg protein. The values of treatment group were always normalized to that of control group.

Statistics
The data were presented as mean ± standard deviation (SD). Statistical differences were analyzed by one-way ANOVA with post hoc Bonferroni test (SPSS version 18.0). Values of p< 0.05 were considered statistically significant.

AntagomiR-451 expression attenuates OGD-induced necrosis of HUVECs
First, HUVECs were cultured with OGD for 3 hours(see Methods), followed by re-oxygenation for another 24 hours. qRT-PCR assay results in Fig 1A demonstrated that miR-451 level in HUVECs was unchanged before and after OGD treatment. Expression of antagomiR-451 [17] caused dramatic decrease of miR-451 in HUVECs (Fig 1A). Significantly, OGD-induced death of HUVECs, evidenced by MTT viability OD reduction ( Fig 1B) and Trypan blue positive cell increase (Fig 1C), was largely attenuated in antagomiR-451-expressing cells. Therefore, miR-451 depletion by antagomiR-451 protected HUVECs from OGD (Fig 1B and 1C). Results in Fig 1D showed that OGD treatment in HUVECs induced significant LDH release to the medium, which is a characteristic marker of cell necrosis. Such effect by OGD was again inhibited by antagomiR-451 (Fig 1D). AntagomiR-C, the control antagomiR, had no significant effect on miR-451expression ( Fig 1A) and OGD-induced HUVEC death (Fig 1B-1D). Using multiple apoptosis assays, including Annexin V FACS assay and TUNEL staining assay, we failed to detect significant apoptosis in OGD-treated HUVECs, which was in line with other studies [3,4,5].Together, these results demonstrate that antagomiR-451 expression depletes miR-451 and attenuates OGD-induced HUVEC necrosis.

Fig 1. AntagomiR-451 expression attenuates OGD-induced necrosis of HUVECs. Parental HUVECs
("Parental") as well as HUVECs expressing antagomiR-451 or antagomiR-C were maintained under OGD for 3 hours, followed by 24 hours of re-oxygenation;miR-451 expression was tested by qRT-PCR assay (A); Cell survival was tested by MTT assay (B); Cell death was examined by Trypan blue staining assay (C); LDH release in conditional medium was also measured as indicator of cell necrosis (D). "Mock" stands for normal culture condition (Same for all figures). "OGD" stands for OGD/re-oxygenation (Same for all figures). Bars indicate mean ± standard deviation (SD, n = 5). *p<0.05 vs. "Mock". # p<0.05 vs. "OGD" of "antagomiR-C" cells. Each experiment was repeated five times and similar results were obtained. silenced cells (Fig 2C and 2D).Thus, antagomiR-451 was in-effective against OGD when AMPKα was silenced (Fig 2C and 2D).These results suggest that activation of AMPK is required for antagomiR-451-medaitd anti-OGD activity. Notably, HUVECs with AMPKα shRNA were more vulnerable to OGD (Fig 2C and 2D). These results imply that basal AMPK activation should also be pro-survival against OGD.

Fig 2. AMPKα knockdown abolishes antagomiR-451-medaitd HUVEC protection again OGD. Parental
HUVECs ("Parental") as well as HUVECs expressing antagomiR-451 or antagomiR-C were maintained under OGD for 3 hours, followed by 24 hours of re-oxygenation, p-/total AMPKα and ACC were shown, blot data were quantified (A).HUVECs, with/outantagomiR-451, were infected with lentiviral scramble control shRNA ("SCR shRNA") or AMPKα shRNA ("s1/s2"); Cells were then subjected to the OGD for 3 hours, followed by 24 hours of re-oxygenation; Expressions of p-/total AMPKα and ACC were shown, blot data were quantified (B); Cell survival and necrosis were tested by MTT assay (C) and LDH release assay (D), respectively. Bars indicate mean ± standard deviation (SD, n = 5).*p<0.05 vs."Parental" cells (A).*p<0.05 vs. group "SCR shRNA" (B).*p<0.05 vs. group "Mock" (C and D). # p<0.05 vs."SCR shRNA" group(C and D).Each experiment was repeated four times and similar results were obtained. HUVECs with antagomiR-451 were unable to resistant to OGD when AMPKα was in-active (Fig 3B and 3C). These results again suggest that activation of AMPK is indeed indispensable for antagomiR-451-mediated cytoprotection in HUVECs. It should be noted that OGDinduced death of HUVECs was again aggravated with DN-AMPKα expression (Fig 3B and  3C), once again confirming that basal AMPK activation is pro-survival in HUVECs.

Discussion
The potential activity of miR-451 in various cancer cells has been extensively studied [38,39,40,41]. It has been shown that miR-451 level was decreased in human cancer, which serves as a potential tumor suppresser [38,39,40,41]. Over-expression of miR-451 could inhibit cancer cell survival and proliferation via silencing different oncogenic proteins [38,39,40,41,42]. In the current study, we showed that miR-451 expression potentiated OGD damages to HUVECs. On the other hand, antagomiR-451 expression inhibited OGD-induced HUVEC necrosis by activating AMPK signaling.
We showed that activation of AMPK was required for antagomiR-451-mediated HUVEC cytoprotection. AMPK inhibition, by AMPKα shRNA or dominant negative mutation, almost abolishedantagomiR-451-medaitd HUVEC protection again OGD. On the other hand, forced activation of AMPK by exogenous expression of ca-AMPKα inhibited OGDinduced death of HUVECs. Remarkably, pre-miR-451 expression led to further AMPK inhibition, thus potentiating death of HUVECs by OGD. Thus, AMPK is indeed pro-survival again OGD in HUVECs. Expression antagomiR-451 activated AMPK to inhibit ODGinduced necrosis of HUVECs.
The pivotal role of p53 in mediating cell apoptosis has been well-established. Very recent studies have also implied the requirement of p53 in promoting non-apoptotic cell death, known as programmed necrosis [28]. Under OGD, activated p53 is shown to trigger mitochondrial permeability transition pore (mPTP) opening via physically interact with Cyp-D in mitochondria, thus inducing mitochondrial depolarization, ROS production and cell necrosis (but not apoptosis) [3,28]. Inhibition or silence this complex was shown to attenuate or even reverse OGD-induced programmed necrosis [3,4,5,28]. In line with these studies, OGD also provoked mitochondrial necrosis pathway in HUVECs, evidenced by mitochondrial p53-Cyp-D association, mitochondrial depolarization and profound ROS production. Our results here demonstrated thatantagomiR-451expression in HUVECs largely inhibited OGD-induced programmed necrosis, which might explain the superior pro-survival activity of antagomiR-451 in HUVECs.

Fig 5. AntagomiR-451 attenuates OGD-induced programmed necrosis in HUVECs.
HUVECs expressing antagomiR-451 or antagomiR-C were maintained under OGD for 3 hours, followed by re-oxygenation for indicated time, mitochondrial fraction was isolated, and the association between Cyp-D and p53 was tested by the coimmunoprecipitation assay ("Mito-IP") assay (A, left panel). "Input" tested Cyp-D and p53 expression (A, right panel); Mitochondrial depolarization was tested by JC-10 assay (B); Relative ROS content (C) and lipid peroxidation level (D) were also tested. Bars indicate mean ± standard deviation (SD, n = 5).*p<0.05 vs. group "Mock". # p<0.05 vs. "OGD" of "antagomiR-C" cells. Each experiment was repeated three times and similar results were obtained.